Development and production of high strength pipeline · PDF fileDevelopment and production of...

TP43

Development and productionof high strength pipeline steels

Dr. Ing. Hans-Georg Hillenbrand EUROPIPE GmbHDr. Ing. Michael Gräf EUROPIPE GmbHDr. Ing. Christoph Kalwa EUROPIPE GmbH

Niobium 2001December 02-05, 2001, Orlando, Florida, USA

Niobium 2001, Orlando, USA 1

DEVELOPMENT AND PRODUCTION OF HIGH STRENGTH PIPELINE STEELSDr.-Ing. Hans-Georg Hillenbrand,

Dr.-Ing. Michael Gräf, Dr.-Ing. Christoph Kalwa

EUROPIPE GmbHFormerstr. 49

40878 RatingenGermany

Abstract

In view of the ever-increasing pipeline length and operating pressure, the development of high-strengthsteels makes a significant contribution to pipeline project cost reduction. In the case of offshorepipelines, the operating pressure of the pipeline does not play the dominating role. It is rather theambient water pressure that is more important. Therefore, one of the design criteria for offshorepipelines is less the strength but more the collapse behaviour of the pipe. The pipe to be used inoffshore pipeline construction should possess not only good materials properties but also goodgeometry to ensure good collapse strength. As the H 2S content of the gas being transported increases,the requirements for HIC resistance of the pipe material increase. When an aqueous phase is present,CO2, H 2S and chlorides are extremely corrosive. For applications in such corrosive environments,either a pipe made of all-corrosion resistant material or a low-alloy steel pipe clad with a high-alloycorrosion resistant material is used.

The initial part of the paper discusses the metallurgical principles and the development of large-diameter linepipe steels. In the second part the production results of different orders represent the state-of-the-art of production of longitudinally-welded large-diameter pipe. Projects of high strength linepipe, pipe for deepwater application, HIC resistant and clad pipe applications are presented.

The paper concludes with the need of close collaboration between all parties involved to optimise thepipeline projects in terms of quality and costs.


Introduction

Over the past 30 years, severe demands have been placed on the pipe manufacturer with respectto the development and processing of materials to linepipe. Generally, longitudinally weldedlarge-diameter linepipe is used for the transportation of oil and gas, because it offers the highestsafety in pipeline operation and represents the most economic solution. From the point of view ofpipeline economy, the pipe must favourably respond to laying in the field and permit highoperating pressures for the pipeline. These requirements imply that the pipeline steel has topossess high strength and toughness and that the pipe shall have optimised geometry.

X 100

X 80

X 70

X 60

X 52

1965 1970 1975 1980 1985 1990 1995 2000

TM + Acc.Cooling.

TM + Acc.Cooling.

TM -treatment

Hot rolledand normalized

0.08 CNb Ti

0.08 C 0.2 MoNb Ti

0.12 CNb V

0.20 CV

API grade

Figure 1: Development of High Strength Steels

The development of high strength steels is shown in Figure 1. In the seventies, the hot rolling andnormalising was replaced by thermomechanical rolling. The latter process enables materials up toX70 to be produced from steels that are microalloyed with niobium and vanadium and have areduced carbon content. An improved processing method, consisting of thermomechanical rollingplus subsequent accelerated cooling, emerged in the eighties. By this method, it has becomepossible to produce higher strength materials like X80, having a further reduced carbon contentand thereby excellent field weldability. Additions of molybdenum, copper and nickel enable thestrength level to be raised to that of grade X100, when the steel is processed to plate bythermomechanical rolling plus modified accelerated cooling.

About 8 million tonnes of linepipe is produced every year. The major part of it consists of pipe instandard material grades, which are not addressed in detail in this paper. The big challenge to thelinepipe manufacturer is posed by the projects that can only be executed by means of specialefforts and measures, that means high strength, offshore, HIC resistance and clad pipe. High-strength pipe in grades X70 and X80 are currently being used in the construction of long-distancepipelines, and pipe in grades X90 and X100 are currently being evaluated. Having explored themajor part of the reservoirs in shallow waters, the drilling activities and hence pipelineinstallations are being shifted into increasingly deep waters. For instance, an offshore pipeline iscurrently being installed in more than 2000 m deep waters. The pipe used in the construction ofthis pipeline has very little in common with the pipe used in the construction of onshore gastransmission pipelines. Additionally good resistance against sourgas is required for the pipes. Inthe case of high corrosive fluids another manufacturing process, cladding of the pipe, has to beestablished.


Pipe with such a combination of partly conflicting properties can only be produced when themetallurgical principles are properly understood and optimally utilised.

Microstructural Influences

Ferritic and Bainitic Structures

Microstructural features such as dislocations, grain boundaries and precipitation, govern themechanical properties of steels. In low-alloy steels, they develop in the course of transformationof the austenite during cooling, and the development depends on the cooling rate and cooling stoptemperature.

X 80Acc. Cooling

X 70 TMT

X 60Normalized

0Transition temperature

Decrease Increase

DH

PH

GRPR

DHPH

GRPR

GR = Grain refinementPH = Precipitation hardeningDH = Dislocation hardeningPR = Pearlite reduction

Figure 2: Microstructural effects for enhancing strength and toughnessproperties

Figure 2 shows, how the combination of the different types of microstructures contribute toincrease mechanical strength and toughness of steels starting from normalised X60 grade, whichwas mainly used in the early seventies (1). The steel typically contains about 0.2% carbon, 1.55%manganese, 0.12% vanadium, 0.03 % niobium and 0.02% nitrogen.

The thermomechanically processed X70 steel mentioned in the figure was microalloyed andcontains only just 0.12% carbon. Thermomechanical rolling results in a significant reduction ofthe ferrite grain size. Grain refinement is the only method by which both strength and toughnesscan simultaneously be improved. The loss of strength resulting from reduced pearlite contentscan be offset by precipitation hardening and dislocation hardening. Reduction of pearlite content,grain refining, dislocation hardening and precipitation hardening contributed individually and incombination to the development of X70 steel with improved weldability and favourable ductile-brittle transition temperatures.

Further increases in strength and toughness, which led to the development of X80 steel, can onlybe attained by changing the microstructure of the steel matrix from ferrite-pearlite to ferrite-bainite. In comparison with the thermomechanically rolled X70 steel, the X80 steel has a further


reduced carbon content, reduced grain size and an increased dislocation density. These two steelgrades also differ in their precipitation characteristics.

X60 NormalizedASTM 7/8

X70 TMTASTM 10/11

X80 Acc.CoolingASTM 12/13

Figure 3: Microstructures of normalised, thermomechanically treated andaccelerated cooled steels

Figure 3 shows typical microstructures of three types of linepipe steel. Banded ferrite and pearliteand coarse ferrite grain size (ASTM 7–8) are the characteristic features of conventionally rolledand normalised X60 steels. The microstructure of TM rolled X70 steels is more uniform and theferrite grains are finer (ASTM 10–11). The most uniform and extremely fine microstructure isattained by accelerated cooling that follows thermomechanical rolling, as shown for the X80steel. The improved properties of this steel can be attributed to its ferritic-bainitic microstructure.

Ferrite Bainite1 µm 1 µm

Figure 4: TEM microstructures of ferrite and bainite

The basic morphological difference between the polygonal ferrite and bainite is illustrated in thefollowing figures. Figure 4 shows electron microscope photographs of the grain structure of thesetwo microstructural types. The effective grain size of bainitic microstructure cannot beestablished by optical microscopy, because, large and small angle grain boundaries are notdiscernible with an optical microscope. Therefore, the electron diffraction patterns of a sufficientnumber of test points on the screen must be systematically examined by means of dark fieldimages (2). As shown in figure 5, a statistical mean grain size of less than 1 µm can be obtainedfor the bainite, whereas the grain size of ferrite is a multiple of this value (3).


0.2 0.5 1 3 5 10 20Grain size d [µm]

Fre

quen

cy

Bainite Ferrite

Logarithmic normal distribution

Figure 5: Grain sizes of ferrite and bainite

A further important difference is the substantially higher dislocation density in bainitic structures.Dislocation density measurements involve considerable experimental expenditure in the electronmicroscope, because in addition to counting out the dislocations, the foil thickness must bedetermined at many points by means of convergent diffraction. The statistical evaluation of suchdislocation density measurements can be seen in figure 6.

0.3 0.5 1 3 5 10 30Dislocation Densityϑ [x109 cm-2]

Ferrite Bainite

Fre

quen

cy

Figure 6: Dislocation densities of ferrite and bainite

The results for the ferritic and bainitic microstructures are based on the measurements on large-diameter pipe made of manganese-niobium steels. The thermomechanical rolling conditions werethe same for both microstructures. The most important micro-structural differences betweenbainite and ferrite are mainly attributable to the lower temperature of formation of bainite.Transformation into bainite can be effected by additions of boron and/or nickel and molybdenumto the steel. Figure 7 shows the continuous cooling transformation (CCT) diagram of a steelcontaining 0.08% carbon, 1.44% manganese, 2.31% nickel, 0.2% molybdenum, and 0.04%niobium, which after cooling in air contains about 50% bainite in the microstructure (4). But


alloy additions made to obtain bainitic fractions increase the carbon equivalent , which may affectthe steel’s field weldability.

Time [sec]100 101 102 103 104

1000

800

600

400

200

0

Temperature[oC]

Chemicalcomposition

[%] .005Mn

.08 .26 1.44 2.31 .20 .040C Si S Ni Mo Nb

Martensite

Austenite

BainiteFerrite

Aircooled

— Ac3 —— Ac1—

50% Ferrite

50% Bainte

Figure 7: CCT-diagram of MnNiMoNb-steels

Time [sec]100 101 102 103 104

1000

800

600

400

200

0

Temperature[oC]

Chemicalcomposition

[%] .003Mn

.11 .34 1.51 .029C Si S Nb

Martensite

Austenite

Bainite

Ferrite

Aircooled

— Ac3 —— Ac1—

50% Ferrite

50% Bainte

Pearlite

TM-Acc.Cool.

Figure 8: CCT-diagram of accelerated cooled MnNb-steels

As can be seen in figure 8, a microstructure consisting of 50% ferrite and 50% bainite can also beobtained with a normal manganese-niobium steel by subjecting it to accelerated cooling. Using aspecial water cooling system at the end of the thermomechanical rolling, the austenite is made topass through the ferritic range in the CCT diagram more quickly, and the formation of pearlite iscompletely suppressed (5).

Effect of Microalloying Elements

In the following, the effect of niobium and titanium on the steel’s microstructure is discussed (3).In case of thermomechanically rolled large-diameter pipe steels, the effect of the microalloyingelements on the mechanical properties is governed by their tendency to bind carbon and nitrogen.


For the purposeful development of micro-alloyed steels, thorough knowledge of the bondingbehaviour is inevitable.

According to the electron microscope photographs of extraction replicas shown in figure 9, thelower final rolling temperature has led to a distinct increase in the population and fineness of theprecipitated niobium carbonitrides. In the case of the titanium carbonitrides, a change in size andquantity – compared with the higher final rolling temperature – is not established.

TEM (extraction replica)

TE = 900°C TE = 795°C

Figure 9: TEM microstructures of carbonitride precipitations

For a number of micro-alloyed steels with varying niobium and titanium contents, the chemicalcomposition of the carbonitride precipitates was determined by means of EDX-analyses andlattice spacing measurements. A substantial proportion of the niobium contained in manganese-niobium steels is used up in the upper austenite range in binding nitrogen. This means that in thelower austenite and/or ferrite range, only a small amount of niobium is available for precipitation.

In a manganese-niobium-titanium steel with a titanium to nitrogen ratio not below thestoichiometric value, the precipitation begins with titanium nitride containing also very lowamounts of niobium and carbon, even at high temperatures. Niobium can then precipitate, mainlyin the form of carbides. The dissolution temperature of carbides is lowered significantly in theabsence of residual nitrogen. Consequently, when the slab is reheated to rolling temperature asubstantial proportion of the niobium gets dissolved, while stable titanium nitrides remain un-dissolved. During thermomechanical rolling in the lower austenite region, fine particles with highniobium contents precipitate anew. This leads to a fine austenite grain size and thus contributes tohardening of the ferrite. On the other hand, a sufficient amount of niobium is available forprecipitation hardening through coherent precipitates in the ferrite.

From a large number of such particle analyses and the pertinent diffraction patterns, the chemicalcomposition of the incoherent precipitates was determined. Figures 10 and 11 show the frequencydistribution of the lattice constants of the carbonitride precipitates in manganese-niobium andmanganese-niobium-titanium steels (6). A comparison of figures 10 and 11 reveals that in thelatter steel, as a result of titanium precipitation, a great proportion of the niobium-carbonitridesexhibits very high carbon contents which at low temperatures are available for the precipitationwith the associated desirable effect.


Number of diffraction pattern30

.420

25

20

15

10

5

0.425 .430 .435 .440 .445 .450

Lattice constant (nm)

TiN TiC

NbN NbC

Figure 10: Distribution of lattice constants of carbonitrides in MnNb-steelsNumber of diffraction pattern

.420

25

20

15

10

5

0.425 .430 .435 .440 .445 .450

Lattice constant (nm)

TiN TiC

NbN NbC

Figure 11: Distribution of lattice constants of carbonitrides in MnNbTi-steels

These microstructural changes can be clearly studied using STEM techniques. Figure 12 gives anexample of EDX analysis of the carbonitride precipitates in a manganese-niobium-titanium steel.The spectrum in the upper part of figure 12 shows the metal contents of the cube-shapedcarbonitrides, containing predominantly titanium, with small amounts of niobium. The lower partof this figure shows the metal contents of the round shaped carbonitrides, which almostexclusively consist of niobium.


Figure 12: EDX-analysis of carbonitride precipitations

Thermomechanical Rolling and Accelerated Cooling

In order to achieve a homogeneous fine-grained microstructure and hence improved strength,good toughness properties and high HIC resistance, compared to steels produced by conventionalthermomechanical rolling, the accelerated cooling process is adopted in the plate mill.

Cooling stop temperature (CST)TM - Acc. Cooling 2

Finish rolling temperature 2

TM - Acc. Cooling 1

Finish rollingtemperature 1 (FRT 1)

Hotrolling

Dissolution of precipitatesand graingrowth Slab reheating temperature (SRT)

Deformedaustenite grains

Recrystallized austenite grains(smaller in size)

Deformedaustenite grains

Deformedaustenite andferritegrainsHeating

Time

Tem

pera

ture

Figure 13: Schematic illustration of thermomechanical rolling with andwithout accelerated cooling during the 2nd and 3rd rolling stage

The metallurgical processes occurring during thermomechanical rolling in conjunction withaccelerated cooling can be understood from the schematic diagram presented in figure 13 (7), inwhich the most important rolling stages and rolling parameters to be controlled are shown.

The cooling system used here can be put into operation twice during rolling. Cooling operation 1enhances the grain refinement of ferrite, whereas cooling operation 2 prevents the formation ofpearlite during cooling, thereby improving the homogeneity of the final microstructure. Important


variables of the cooling operations are cooling rate and cooling stop temperature. This specialcooling system permits cooling to be carried out after the second and third stages of rolling sinceit is firstly, quite compact and, secondly, installed close to the rolling stand.

The essential rolling parameters of the thermomechanical process are:- the slab reheating temperature (SRT) for dissolution of the precipitated carbonitrides,- the roughing phase for producing a fine, polygonal austenitic grain by means of

recrystallisation,- the final rolling temperature (FRT), which must be maintained within the range of the non-

recrystallising austenite, and- the degree of final deformation (FD) in this temperature range.

If accelerated cooling is employed, the following additional parameters shall be considered:- the cooling rate, and- the cooling stop temperature (CST).

0.04C – 1.3Mn – 0.04Nb

TM treatment TM treatment + Acc.Cooling1 and 2

Figure 14: Effect of accelerated cooling on the microstructure of TM-steels

Figure 14 shows the effect of accelerated cooling that follows the first and the second finishrolling operations on microstructure. A few pearlite islands are present in the mid-wall region ofthe thermomechanically rolled plate. As a result of the two-stage cooling, not only the ferritegrain size is further refined but also the pearlite is replaced with bainite. The microstructure of theaccelerated-cooled material gives the impression of homogeneity. The homogeneity of thismicrostructure has also found expression in the strength and toughness properties of theaccelerated cooled material.


High Strength Linepipes

Over the past two decades, extensive work has been carried out to develop high-strength steels ingrades X80 and X100 to assist customers in their endeavour to reduce weight and pipelayingcosts.

Development to X80

The developments in alloy design since 1984 can be seen in the results (8) on pipe producedcommercially for the Megal II, CSSR and Ruhrgas projects (Table I).

A manganese-niobium-titanium steel, additionally alloyed with copper and nickel, was used inthe production of the 13.6 mm wall pipe for the Megal II order in 1984. Subsequent optimisationof production parameters enabled the CSSR order to be executed using a manganese-niobium-titanium steel without the additions of copper and nickel. This has simultaneously led to animprovement (i.e. reduction) in the carbon equivalent of the steel used.

Table I: Chemical Composition of GRS 550 / X80 pipe

PCM

.206

.197

.213

Mo

.01

.01

.01

Cu

.18

.03

.03

N

.0052

.0060

.0040

Ni

.18

.03

.03

Cr

.04

.05

.05

S

.0016

.0017

.0011

C

.081

.085

.09

Megal II

CSSR

Ruhrgas

44" x13.6mm

56" x15.6mm

48" x18.3mm

Pipegeo-metry

Order Si

.42

.38

.40

Mn

1.89

1.85

1.94

P

0.011

0.014

0.018

Al

.038

.030

.038

Nb

.044

.044

.043

Ti

.018

.019

.017

IIW

.430

.409

.435

In 1992, 48” diameter pipe for a Ruhrgas pipeline project in Germany with 250 km lengthrequiring GRS 550 (X80) was produced. Also pipes of 48” diameter with 19.3 mm wall thicknesswere included. Since the strength decreases as the wall thickness increases, it was necessary toraise the carbon and manganese levels marginally. The concentrations of all other elementsremained unchanged.

Table II: Mechanical Properties of GRS 550 / X80 pipe

603

607

592

Megal II

CSSR

Ruhrgas

44" x13.6mm

56" x15.6mm

48" x18.3mm

Pipegeo-metry

Order

Yieldstrength

MPa

Tensilestrength

MPa

Elongation5d (%)

CVN, -20 oC(Joule)

X S X S X S X S

21

18

14

737

716

729

22

10

16

22

23

21

2

2

1

182

183

176

30

29

32

DWTT (-20°C) SA >85%Table II shows the mean values of the mechanical properties measured on the pipes produced forthe three orders. The measured tensile and the impact energy values conformed fully to thespecification requirements in all cases. The standard deviation for the yield and tensile strengthvalues is very low. The impact energies measured on Charpy V-notch impact specimens at -20°C


is very high, resulting in an average value of about 180 J. The 85% shear transition temperaturesdetermined in the drop weight tear (DWT) tests are far below –20°C.

Figure 15 gives an idea of the Ruhrgas pipeline project (9). After welding, non destructiveexamination and coating, the girth welded segments were lowered onto the prepared trenchbottom.

Werne

Schlüchtern

Essen

Bonn

BadBerleburg

Arnsberg

Wetter

Lauter-bach

Vitzeroda

Erfurt

Unna

Frankfurt

Figure 15: Pipe laying of the Ruhrgas Schlüchtern-Werne-ProjectBase Material - Tensile Test

YieldStrength

Elongation / %

560 570 580 590 600 610 620 630 640 650 660 670 680 690 700 710 720 730 740

35302520151050

4035302520151050

Yield/Tensile Strength / MPa

TensileStrength

0.76 0.80 0.90 28 29 30 32 33 34 3531

Frequency of Values / %


25

20

15

10

5

0

Rx / Rm0.78 0.82 0.84 0.86 0.88


Figure 16: Distribution of mechanical properties of 48”diam. X80 linepipewith 15.1 mm wt

One of the latest projects using X80 was a UK pipeline for BG Transco in 2000. The results ofthe tensile tests, performed in the context of certfication of the pipe, are shown in figure 16. Allvalues conform to the requirements of X80. The standard deviations are 15 MPa for yieldstrength, 13 MPa for tensile strength, 0.02 for yield to tensile ratio and 1.8% for elongation.

Table III contains the results obtained on commercially produced 36” diameter pipe with32.0 mm wall thickness in API grade X80. The manganese-niobium-titanium steel used here hasa sufficiently high ratio of titanium to nitrogen and is additionally alloyed with molybdenum. The


low carbon equivalent ensures good field weldability. The elongation values (A2”) areparticularly high. The Charpy V-notch impact energy measured at -40°C is in excess of 200 J andthe shear area of DWTT specimens tested at –20°C is greater than 85%. The forming andwelding operations carried out on this high strength steel did not cause any problems.

Table III: Chemical and mechanical properties of 36” diam. X80 pipe with32 mm wt

Transverse(flat bar)

Transverse(round bar)

MPa

Rt0.5Charpy V-notch (1/1),

transverse, -40oC DWTT, -20oC

Aver.J

Rm Rt/Rm

MPa % %

A2"

J J J S.A.%

S.A.%

559

579

685

674

82

86

47

46

222 219 231 224 85 90

Specimenorientat .

(and type)

S

.0010

C

.07

Si

.27

Mn

1.86

P

.015

Al

.036

Cu

.02

Cr

.03

Ni

.02

Mo

.15

Nb

.040

Ti

.023

N

.0057

IIW

.419

PCM

.186

Mean chemical composition [wt.%]

Development to X100

No technological breakthroughs, such as TM rolling and accelerated cooling which increased thestrength and toughness respectively, but only improvements in the existing technology wereinvolved in the production of grade X100 plate. As a result, the production window is quitenarrow. Heat treatment of plate or pipe is obviously not advisable.

As can be seen in figure 17, three different approaches are generally possible with respect to theselection of chemical composition and cooling conditions (10).

Approach A, which involves a relatively high carbon content, has the disadvantage that the crackarrest toughness requirements to prevent long-running cracks, may not be fulfilled. Moreover,this approach is also detrimental, e.g. to field weldability.

Approach B, which is practically the same as direct quenching with fast cooling rates down to avery low cooling-stop temperature, results in the formation of uncontrolled fractions ofmartensite in the microstructure, which have a detrimental effect on toughness properties of basemetal and leads additionally to the softening in the heat affected zone. This effect cannot beadequately compensated for extremely low carbon contents, without adversely affectingproductivity. Moreover, it is very difficult to produce pipe with adequate uniformity of strengthproperties. This difficulty cannot be attributed solely to the Bauschinger effect associated withthe variation in local deformation occurring during the heavy straightening operation needed inthe case of thin section plate, which distorts heavily during direct quenching.


Approach A: High Carbon, Medium CE, Soft AcCApproach B: Low Carbon , Low CE, Direct quenching (Strong AcC )Approach C: Medium Carbon, Medium CE, Medium AcC

A

B

C

Steelchemistry

Coolingparameters

C-content CEIIW

Cooling rate(AcC)

Cooling stoptemperature (AcC)

0.08%

0.06%

0.06% 0.43

0.48

0.49

high

highlow

low

Figure 17: Modification of steel chemistry and cooling parameters to achievethe strength level of X100

Experience gained meanwhile indicates that Approach C is the best choice. This approachenables the desired property profile to be achieved through an optimised two-stage rollingprocess in conjunction with a reduced carbon content, a relatively high carbon equivalent andoptimised cooling conditions. The special potential of the existing rolling and cooling facilitiescontributes significantly to the success of this approach.

Approach C, which involves a low carbon content, ensures excellent toughness as well as fullysatisfactory field weldability, despite the relatively high carbon equivalent of the chemicalcomposition. The chemical composition should therefore be considered acceptable for thepurpose of current standardisation.

Figure 18: Full-scale burst tests on X100

Large-diameter grade X100 pipes were produced on a trial basis for the third and fourth time. Thepipes are intended for use as test pipes to determine the behaviour of grade X100 material in full-


scale burst tests (figure 18), which are conducted as part of an ECSC-funded research project(11).

The mechanical properties determined on the pipes are quite appreciable, although they do notfully conform to the requirements currently specified for lower-strength material grades. Thecurrent requirements particularly for yield to tensile ratio and elongation, which are applicable tomaterial grades up to X80, are too severe to be fulfilled for grade X100 on a statistical basis. Thisdifficulty is attributable to the basic finding that as the strength increases, the yield to tensile ratioincreases and the elongation decreases. The X100 material produced responds favourably tomanual and mechanised field welding, a finding which can be attributed to its reduced carboncontent.

For reasons of technical feasibility and cost-effective production, it is necessary in the context ofgrade X100 to reassess and redefine some of the requirements for the mechanical properties,considering the anticipated service conditions for the pipe.

Linepipes for Deepwater Application

A further new challenge to the pipe manufacturer and the pipelaying contractor is the offshorepipeline in deeper waters. A consortium has been carrying out a feasibility study since 1993, for agas transmission pipeline from Oman to India in 3500 m deep water.

Table IV shows the most important requirements for the pipe (12). To prevent collapse of thepipeline under an ambient external pressure of about 350 bar, the linepipe to be used has to meetsevere requirements. The extreme pipe geometry was characterised by a wall thickness of 41 mmand 28” diameter.

Table IV: Requirements for Oman-India gas-transmission pipelineLength of submarine line: 1200 kmMax. water depth: 3500 mMax. pipe size: 610 mm I.D. x 41 mm W.T.

Material: X70 non-sour

Yield strength, 482 - 586 MPaTensile strength, 565 - 793 MPaCVN base, -10oC: 200 / 150 JCVN weld, -10oC: 100 / 75 JDWT, -10oC: min. 85% SACTOD weld, -10oC: min. 0.15 mm

Out - of - roundess: ≤ 4 mm

} Long.-andtransverse

Mechanical strength and geometry of the pipe are the two most important factors that affect thecollapse strength. High strength values, which need to be uniform over the pipe circumference,reduce the susceptibility of pipe to collapsing. The material for the pipe is grade X70 (non-sour).For offshore pipelines, the pipe is often required to meet the requirement for yield strength also in


the longitudinal direction. The toughness requirements are quite high at >200 J average and>150 J individual for the base material and >100 J average and >75 J individual for the weld.DWT test and CTOD test are also specified. Out-of-roundness of the pipe has a detrimental effecton collapse strength. Hence, the requirement for out-of-roundness reads: as low as possible, but4 mm maximum.

baseweld

C0.090.07

Si0.250.29

Mn1.691.40

P0.0110.012

S8 *)41 *)

Al0.0440.022

Cu0.020.03

Cr0.030.04

Ni0.220.15

baseweld

Mo0.010.20

V0.080.05

Ti0.0030.022

Nb0.050.03

N36 *)56 *)

B1 *)

36 *)

CE0.410.38

PCM0.200.20

Chemical composition (wt%): Base material and weld metal

*): ppm

Figure 19: Chemical composition and macrosection of Oman-to-India-X70 pipes,28” diam. with 41.0 mm wt

An exclusive contract was awarded in 1995 to produce the first 1000 m of pipe for weldabilitytrials. Forming pipe with such an unfavourable diameter-to-thickness ratio places severe demandson the pipe forming equipment, e.g. the crimping press, U-ing press, O-ing press and mechanicalexpander. Extensive laboratory work involving finite element analysis was conducted todetermine the parameters needed for forming the pipe and the loads occurring on the formingequipment. Figure 19 shows the chemical composition and the macrosection of the linepipesproduced for this contract.

The variation in pipe strength, the pipe geometry and the Bauschinger effect caused by pipeforming operation all have an effect on the collapse strength. A further aspect is the ovality thatshould be less than 4 mm; results of the produced linepipes are shown in figure 20.

Some collapse tests on linepipe were performed to investigate the collapse behaviour (figure 21).The results are given in table V. The values shown as calculated collapse pressures weredetermined by an analytical procedure (figure 22). The calculated values are greatly dependent onthe compressive yield strength, which can be determined experimentally, and on the initialovality.


50

40

30

20

10

00 1 2 3 4

Ovality [mm]

Frequency [%]

5

South endNorth end

O.D. 711 mm x 41.0 mm w.t.

Figure 20: Production results of Oman-to-India X70 pipes ovality at pipe ends

Figure 21: Photograph showing buckled linepipe section

As can be seen, there is nearly no difference in buckling pressure between pipes of differentstrength for a given level of ovality, except in one case. The effect of compressive yield strengthof the material on the buckling pressure is quite significant in the case of pipes with a thicker wall(13).

Furthermore, it is remarkable that the 28” diameter pipe with 41.0 mm wall thickness subjected toa simulated coating thermal cycle has a higher collapse pressure than does the pipe of the samediameter not subjected to the simulated thermal cycle, but with a higher wall thickness of44.0 mm. This result can also be explained in terms of the Bauschinger effect. As a result of thethermal treatment (at 200°C) prior to testing, the Bauschinger effect on the mechanical strengthof the pipe material was almost completely eliminated.


Table V: Results of collapse tests

36" dia. X 17.5mm wt(X65)36" dia. X 31.0mm wt(X65)28" dia. X 35.6mm wt(X65)28" dia. X 42.0mm wt(X65)28" dia. X 41.0mm wt(X70 coating .simul.)28" dia. X 44.0mm wt(X70)

Measuredcollapsepressure

(bar)

Calculatedcollapsepressure

(bar)

OvalityUo (%)

Compress .yield

strengthσc0.2(MPa)

Tensileyield

strengthRt0.5

(MPa)

Pipe

31.0

147

311

421

484

468

30.8

154

359

427

475

459

0.5

0.4

0.5

0.7

1.4

0.6

400

425

410

425

525

435

470

490

485

490

540

550

Figure 22: Simulation of collapse behaviour with FEA

Linepipes for Sour Service

The development of sour service grades for linepipe with resistance to hydrogen inducedcracking (HIC) over the past 10 years has been considerably influenced by the requirements ofthe market. In the following, the state-of-the-art of linepipe production is illustrated using thedata on commercially produced pipe.

The new linepipe steels for sour service are characterised by extremely good toughness propertiesand high resistance to HIC. These properties have been achieved by means of lean chemistry andhighest steel cleanness. This ensures that the steels have a low hardenability and the formation of


inclusions and precipitates that are detrimental to HIC is avoided. Figure 23 gives an idea of theefforts the steel-makers make to cope with these requirements.

To meet the requirements for other properties, such as strength, weldability and fabricability, thesteel composition and the process parameters for the steel making and plate rolling have to bedesigned duly. It is important to mention that each production step has to be performed adheringclosely to the parameters, such as time regime for each step, specially established for theproduction of HIC resistant steels.

Steelmaking step Metallurgical objectives

Hot metal Desulphurisationdesulphurisation

Converter process DephosphorisationDecarburisation

Vacuum tank degassing Desulphurisation

Steelmaking step Metallurgical objectives

Cleanness stirring Cleanness improvement

Ca treatment Inclusion shape control

Casting Prevention of resulphurisationMinimisation of segregationAssurance of cleanness

Figure 23: Production steps of steels with high cleanness requirements forsour service

The accelerated cooling process in the plate mill leads to a more homogeneous microstructureresulting in improved HIC resistance. The banded ferritic-pearlitic microstructure is than replacedby a ferritic-bainitic structure (14).

Table VI: Production results on 36” diam. X65 linepipe for sour service260 000 tonnes with 28.4 mm wt

Transverse yieldstrength Rt0.5 [MPa]Tensile strengthRm [MPa]Yield-to-tensile ratioRt0.5 /Rm [%]Elongation A2"[%]CVN toughness, -20oCWeld metal [J]HAZ [J]Base metal [J]DWTT [%SA]

Mechanicalproperties Standard

485

572

85

50.4

8431033693

Mean

14

11

1.94

1.78

2565383.3

Meancomp. .004 .127.286.044.07.038.0008.0121.34.31.04

C Si Mn P S Al V Nb N IIW PCM

Table VI shows the results on 260,000 t of production pipe for the construction of an offshore gastransmission pipeline (15). In the table the mechanical properties are described by means of


average values and standard deviations. The requirements for grade X65 pipe were comfortablymet. The important requirements for the chemical composition of an HIC resistant steel are lowconcentrations of carbon, manganese and sulphur.

As this table reveals these requirements have been strictly observed in the chemical compositionof the pipe used for this X65 offshore project. Besides the low carbon and manganese contentsthe steel features microalloying additions in the way of vanadium and niobium, which were madeto meet the requirements for the mechanical properties. The HIC test results on this order aregiven in table VII. Under the standard test conditions according to NACE (pH = 3, 1bar H 2S ) thespecified acceptance criteria are reliably fulfilled.

Table VII: Typical HIC test results on sour service pipe

Testcondition Base

Result on pipePipe

36“ O.D.x28.4 mmX65 Sour service

Acceptancecriteria Weld

Specification

pH 3CTR: –CLR≤ 10.0%CSR≤ 2.0%

–CLR≤ 10.0%CSR≤ 0.5%

–CLR≤ 10.0%CSR≤ 0.5%

In order to find a correlation between laboratory tests and real corrosion behaviour of thecomponent, a series of full-scale tests were carried out. The laboratory tests were performed inpH 5 and pH 3 solutions for 92 h. The full-scale tests were performed in the pH 3 solution, thetest duration being 90 days. Both HIC-susceptible and HIC-resistant steels were included in thetest. Table VIII shows the results of these tests. Two significant results can be read from thetable. Firstly, the ranking of the materials based on the results of the laboratory tests is confirmedby the full-scale tests. Secondly, the HIC resistant material did not suffer any internal crackingand surface blistering in the full-scale test, although the test solution used was very aggressive.

Table VIII: Laboratory HIC results versus full-scale test results(solution pH3)

LaboratoryHIC test

96 h

Full-scale test90 days(2160 h)

CLR = 75.7%CTR = 12.4%CSR = 3.7%

Heavy crackingand blistering

CLR = 5.7%CTR = 0.4%CSR = 0.0%

No cracksNo blisters

HIC-susceptible steel HIC-resistant steel

ResultsTest

A series of tests was carried out, prior to the manufacture of linepipe for several orders, tooptimise the weld metal composition such that its hardening behaviour is comparable to that ofthe plate material whilst maintaining the required toughness. This has been achieved by using a


specially developed, low carbon, titanium-boron alloyed wire of an otherwise nearly similarcomposition to the parent material. An additional factor that was necessary for fulfilling thetoughness requirements was the limitation of the nitrogen content of the weld metal. For somebig orders, the toughness requirements for weld metal centre and heat-affected zone at -30°Cwere 37 J minimum on individual specimens and 45 J minimum for the mean of three specimens.Despite these exacting testing conditions, all orders could be executed, comfortably fulfilling thespecification requirements. Molybdenum-free titanium-boron-alloyed weld metals as well astitanium-boron-free manganese-nickel-molybdenum weld metals have been used, as required bythe customers.

In order to make the step from a heavy wall grade X65 pipe on to a heavy wall grade X70 pipe(30 mm W.T.), two different approaches have been developed and applied. The idea of increasingcarbon and manganese contents to improve the strength of the steel was abandoned, becausethese elements segregate and enhance centreline segregation, thereby deteriorating the HICresistance of the steel. The Carbon content was 0.04%, the mangenese content 1.35%.

Table IX: Mechanical properties of X70 steels for sour service

Mechanicalproperties

NbV0.08V 0.04NbCuNiMo~ 0.2

NbTi0.08Nb 0.025Ti

Yield strength Rt0.5 [MPa] at RTTransverseLongitudinal

Tensile strength Rm [MPa] at RTTransverseLongitudinal

Yield-to-tensile ratio Rt0.5/Rm [%]TransverseLongitudinal

Elongation A5 [%] at RTTransverseLongitudinal

CVN toughness at -30oCBase metal [J]HAZ [J]Weld metal [J]

DWTT [%SA]Base metal at -10°CBase metal at -20°C

502504

582566

8689

2527

47095

140

9282

509519

595582

8689

2425

480135160

9791

The first approach to increase the strength of the steel aimed at the distribution and type ofmicrostructural constituents and at achieving additional solid solution hardening. The classicalcomposition of the niobium vanadium-type steel, used for grade X65 pipe, was modified byadding or increasing the concentrations of copper, nickel, chromium and molybdenum in the steelanalysis (table IX). This concept results in a carbon equivalent according IIW of 0.39.

Also shown in table IX is the other approach, which is based on increasing the niobiumconcentration and adding titanium to the steel (16). The niobium content is increased to havehigher amounts of niobium in solid solution in the austenite-region as it retards the austenite toferrite transformation and to increase the strengthening by precipitation hardening. Titanium wasadded to bind nitrogen thereby preventing the precipitation of niobium carbonitride and makingniobium more effective for increasing the strength. This approach leads to an carbon equivalentaccording IIW of 0.32.


It should be noted that both the approaches (referred to as NbV and NbTi type in the tables) wereapplied in combination with an improved accelerated cooling process after final rolling toproduce HIC resistant plates with 30 mm wall thickness in grade X70.

These plates were formed by the U-ing – O-ing – Expanding process into 30” diameter pipes.Following the pipe expansion, samples were taken to determine their mechanical properties andcorrosion behaviour.

Table IX also contains the mean values of the mechanical properties for the two differentapproaches. As can be seen from the data the requirement for a shear area of 85 % minimum inthe DWT test –10°C are met by both variants. The Charpy-V-notch impact energy valuesmeasured at –30°C are as expected for a low-carbon steel, at above 450 J. In the weld seam andthe heat affected zone impact energy values in the range of 100 to 160 J were measured. Ingeneral, the niobium vanadium approach yielded more favourable strength values so that thestrength requirements for grade X70 can be met in both transverse and longitudinal directions.

The HIC tests were performed in accordance with NACE standard TM0282-96, solution A(pH = 3) on base material and weld seam. Both approaches fulfilled the requirements for HICresistance as can be seen in table X. The niobium vanadium approach yielded again morefavourable.

Table X: HIC test results of X70 pipe

Specification Requirements NbVNbTi

pH 31bar H2S

Test ConditionAcceptance

Criteria Base Metal Weld Metal Base Metal Weld Metal

CTR < 5%CLR < 15%CSR < 1,5%

< 3%< 8%< 1%

< 5%< 7%< 1%

< 2%< 7%< 1%

< 4%< 6%< 1%

With the steadily increasing demands it becomes progressively difficult to fulfill the standardrequirements for HIC requirements according to the above mentioned standard test conditions. Inthose cases where these requirements can not be consistently achieved, fit-for-purpose testingmethods are currently emerging.

The example given below concerns an inquiry for a linepipe of 42” diameter with 31.5 mm wallthickness in grade X65 or, if practicable, in grade X70 and HIC requirements. A lean chemicalcomposition that is typically used for the production of linepipe intended for sour service couldnot be used here because of the heavy wall. The chemical composition was optimised for mostpart to fulfil the requirements for mechanical properties. Attention was paid to each and everyproduction step with a view to improving HIC resistance of the steel. Different HIC test variantswere tried out to work out a procedure that is close to the predicted service conditions and that isnot difficult to implement. Figure 24 shows the behaviour of the 31.5 mm thick grade X65material, produced on a trial basis, in the various HIC test variants. The HIC index shown on theY-axis is a measure of the extent of cracking. The higher the value of the index, the larger is theextent of cracking. The pH of the test solution was 3. The first row beneath the bars indicates thetypes of specimen used. Standard HIC specimens and large plate specimens were tested. Theplate specimens were hydrogen charged by placing a glass tube on the specimen and filling itwith the test solution. The second row beneath the bars shows the number of specimen sides


exposed to the test solution. Only one-sided hydrogen charging was used in the case of platespecimens. The third row shows the partial pressure of H 2S in the H 2S + N2 gas mixture withwhich the test solution was saturated at atmospheric pressure. The figure clearly demonstrates theeffect of test conditions on the HIC index.

HIC61

HIC6

0.1

Plate11

HIC11

Plate1

0.1

HIC1

0.1

6

5

4

3

2

1

0

HIC Index

Specimen type:Exposed sides:

p (H2S), bar:

Figure 24: HIC behaviour of X65 steel designed for “slightly sour service”

To be on the safer side, the order was executed in grade X65 and the HIC tests were carried outusing 1-side exposure in pH3-solution, but with a H 2S partial pressure of 0.1 bar.

About 200,000 t of pipe was produced for this order. The pipe’s diameter produced was 42” with29.8 mm or 37.9 mm wall thickness. Table XI shows the mean chemical composition and themean mechanical properties determined on the 37.9 mm wall thickness pipe.

Compared to a conventional grade X65 material intended for sour service, the present steel had ahigher carbon content and a higher manganese content. The higher carbon and manganesecontents were intended to ensure that the steel should attain the required mechanical properties,particularly the toughness, at the heavy wall involved. The steel was made to the standard sourservice practice, including vacuum degassing. It was desulphurised to a very low sulphur level of8 ppm average and calcium-treated. This practice was adopted to ensure that the steel meets therequirements for HIC resistance.

As can be seen from the mean values and the standard deviations given in the table, the pipecould be produced with a high statistical confidence level, despite its heavy wall. Both thetransverse and longitudinal tensile properties of the pipe are comfortably above those required forgrade X65. The Charpy V-notch impact energy values measured on the base material at -30°C arein excess of 250 J.


Table XI: Production results on 42” diam. X65 linepipe intended for “slightlysour service” with 37.9 mm wt

Mechanicalproperties Standard deviationMean

Meancomp. –.13.005.00081.60.20.078

C Si Mn P S Al Cu Cr

Yield strength Rt0.5 [MPa]TransverseLongitudinal

Tensile strength Rm [MPa]TransverseLongitudinal

Yield-to-tensile ratio Rt0.5/Rm [%]TransverseLongitudinal

Elongation A2“ [%]TransverseLongitudinal

CVN toughness at -30oCWeld metal [J]HAZ [J]Base metal [J]

.008

Meancomp. .37.004.017.035.27

Ni Mo Nb Ti N IIW PCM

504514

607596

8387

22.122.8

11076

266

2116

1816

2.11.6

1.271.67

373735

– .18

The HIC specimens tested to the fit-for-purpose procedure showed no ultrasonic indications,thereby complying with the specification.

Thus, the steel chemistry selected and the steelmaking practice adopted have proven the rightapproach to execute this special order successfully.

Clad Linepipe

The use of high alloy materials for the linepipe often represents the only solution to combatcorrosion in situations where inhibition and gas processing is not practicable. Clad pipe combinesthe excellent corrosion behaviour of high-alloy austenitic materials such as Incoloy 825 and thehigh strength of carbon steels to an economic solution. Therefore a technology of making cladpipe from clad plate, which can be produced by sandwich-type rolling or by rolling of explosion-clad slabs, was developed. In both cases, a metallurgical bond between the substrate and thecladding is ensured.

In rolling a clad slab into plate, the thermomechanical rolling parameters adopted play animportant role in achieving the strength properties of the substrate material and the corrosionresistance of the clad material (17). Thermomechanical rolling coupled with accelerated coolingis performed in two temperature ranges. The temperature of the first rolling stage is in a rangejust above the austenite recrystallisation regime. Heavy rolling passes, leading to a total reductionof 60% are essential for achieving a fine grained austenitic structure. The finish rolling isperformed above Ar3 temperature (about 800°C) in the austenite non-recrystallisation range todevelop a fine grained transformation product. Additional niobium retards the recrystallisation ofthe austenite. The temperature range between 800 and 500°C must be passed through very fastwith the help of accelerated cooling to avoid the precipitation of any phases that sensitise thecladding material and to achieve high strength and toughness properties for the base material.


Figure 25 gives a rough summary of some details of the longitudinal seam weld. The two-passweld is performed by the multi-wire submerged-arc welding process. Dilution of the inside weldfrom high-alloy cladding material is prevented by adopting a special groove geometry andcontrolling the heat input.

Beadsequence

2

3

Details of welding

Weldingprocess

Weldingconsu -mables1

Passno.

1and

2

3

Multi-wireSAW

Electro -slag

weldi ng

Mn-NiMoor Mn-Mo

(wire)

Inconel625

(strip)

Pipe material

Basematerial

MicroalloyedC-Mn steel

(X 65)

Claddingmaterial

Incoloy 825

Figure 25: Welding consumables and typical build up of the longitudinal weld seamin clad linepipe (24” diam. with 20.5 mm wt)

Table XII: Chemical composition and mechanical properties of large diameter cladlinepipe X65 / Incoloy 825 (24” diam. with 20.5 mm wt)

ToughnesElongation

C0.0350.019

A*)B

Si0.110.24

Mn1.500.81

Cu0.051.40

Ni0.2739.85

Cr0.0623.43

Mo0.013.40

V0.0050.05

Nb0.080.02

Ti0.0150.73

Yield strength (Rp)Tensile strength (Rm)Yield-to-tensile ratio (Rp/Rm)

(A2”) (CVN, -30°C or -22°F)

DWT (85% SA)Shear stress (ASTMA 265)

490 MPa590 MPa

83%42%330 J

< -20°C460 MPa

b) mechanical properties of base material(transverse specimens)

a) Chemical compositions of base material (A) and cladding material (B), wt %

*) PCM = 0.16%

Table XII shows production results on Incoloy 825 clad X65 linepipe (24” diameter x 20.5 mmwall thickness). The substrate steel (X65) has a very good toughness and a low transitiontemperature in the DWT test. The pipe was welded from inside using the electro-slag stripwelding process which leads to very little dilution from the parent material. The niobium contentof the substrate pipe was increased to 0.08 % to ensure adequate strength values after formingand welding.


Conclusion

The predicted growth in energy consumption in the coming decades necessitates severe effortsfor transporting large amounts of natural gas to the end user. Large-diameter pipelines serve asthe best and safest means of transport. This paper gives an overview of the current requirementsof high strength steels, heavy wall pipe for deep sea application, HIC-resistant materials and cladpipe and the associated developments. The technical possibilities were described. Also for thefuture additional substantial improvements can be realised.

The enormous pressure on the price of natural gas forces the pipeline operator to explore all thepossibilities to reduce the cost of a pipeline project in the future. The pipe manufacturer can assisthim in his endeavour by supplying high quality pipe. The effect of pipe quality on project costswill be more substantial when the pipeline is constructed to the limit state design. For an offshorepipeline in the North Sea, the project team could reduce the project costs especially fortransportation in several stages by optimising the pipe diameter, that led to a cost reduction of5%, increasing the material grade, that led to 4%, and using of limit state design, that led to 6%reduction.

Finally, the pipe manufacturer makes his contribution to reducing operational costs of a pipelineover its life by investigating the fatigue, corrosion and ageing behaviour of pipe and pipematerials. These properties have a significant bearing on the integrity of a pipeline andconsequently the operating costs. These properties are currently being extensively studied. Theknowledge gained these studies can be made available to assist the pipeline operator whenplanning a new pipeline project or when estimating the residual life of an ageing pipeline.

For reasons of technical feasibility and cost-effective production, it is necessary in the future toreassess and redefine some of the materials requirements for pipe, considering the anticipatedservice conditions for the pipe. Close co-operation between the pipeline designer, operator, layingcontractor and the pipe manufacturer in this respect would be conducive to finding good solutionsin modern pipeline construction.

References

1. P.A. Peters and H.G. Hillenbrand, “Experience in Supply of Arctic Grade Line Pipe for SovietConstruction Projects”, World Materials Congress 1988, Chicago, USA, 1988

2. M.K. Gräf et al., “Relationship between Microstructure and Mechanical Properties ofThermomechanically Treated Large Diameter Pipe Steels”, 1983 International Conference onTechnology and Application of HSLA Steels, Philadelphia, USA, 1983

3. H.G. Hillenbrand and P. Schwaab, “Determination of the Microstructure of High StrengthStructural Steels for Correlation with their Mechanical Properties”, Mat. Sci. Eng., 94 (1987)

4. M.K. Gräf, H.G. Hillenbrand, and P.A. Peters, “Accelerated Cooling of Plate for High-StrengthLarge-Diameter Pipe”, Conference on Accelerated Cooling of Steels, Pittsburgh, USA, 1985


5. W.M. Hof et al., “New High Strength Large-Diameter Pipe Steels”, CSM Conference, HSLASteels, Beijing, China, 1985

6. H.G. Hillenbrand and P. Schwaab, “Determination of the Bonding Behaviour of Carbon andNitrogen in Microalloyed Structural Steels”, Mikrochimica Acta [Wien], Suppl.11, p. 363-370,1985

7. P.A. Peters et al., “Modern Linepipe Steels for Offshore Use”, Offshore Pipeline TechnologySeminar, Stavanger, Norway, 1988

8. A.W. Gärtner, M.K. Gräf, and H.G. Hillenbrand, “A Producer’s View of Large-Diameter Line-pipe in the Next Decade”, International Conference in Pipeline Reliability, Calgary, Canada, 1992

9. H.G. Hillenbrand et al., “Procedures, Considerations for Welding X80 Linepipe Established”,Oil & Gas Journal, 37 (1995)

10. H.G. Hillenbrand et al., “Development of Large-Diameter Pipe in Grade X100”, PipelineTechnology Conference, Brugge, Belgium 2000

11. G. Demofonti et al., “Large Diameter API X100 Gas Pipelines: ‘Fracture PropagationEvaluation by Full Scale Burst Test’”, Pipeline Technology Conference, Brugge, Belgium 2000

12. M.K. Gräf and G. Vogt, “Experiences with Thick Walled Offshore Pipelines”, Deep OffshoreTechnology, 9th International Conference and Exhibition, The Hague, 1997

13. P.R. Stark and D.S. Mc Keehan, “Hydrostatic Collapse Research in Support of Oman IndiaGas Pipeline”, 27th Annual OTC, Houston, USA, 1995

14. H.G. Hillenbrand, W.M. Hof, and B. Hoh, “Modern Linepipe Steels and the Effect ofAccelerated Cooling”, 4th International Steel Rolling Conference, Deauville, France, 1987

15 M.K. Gräf et al. “Review of the HIC Test Requirements for Linepipe over the Years 1975 to2000”, 3R International, issue 10-11/99, Vulkan Verlag

16. K. Hulka, J.M. Gray and F. Heisterkamp “Metallurgical concept and full-scale testing of a hightoughness, H 2S resistant 0.03 % - 0.10 % Nb steel”, Niobium Tech. Report, NbTR-16/90, 1990

17. M.K. Gräf et al.,“Production of Large Diameter Clad Pipe for the Transportation of HighlyCorrosive Crude Oil and Natural Gas”, 9th Offshore South-East Asia Conference, Singapore,Malaysia, 1992

Development and production of high strength pipeline · PDF fileDevelopment and production of...

Documents

Transcript of Development and production of high strength pipeline · PDF fileDevelopment and production of...